NYU Wireless

Sundeep Rangan, Ted Rappaport, Elza ErkipZoran Latinovic, Mustafa Riza Akdeniz, Yuanpeng Liu

NYU-Poly

June 25, 2013Communications Theory Workshop

Phuket, Thailand

1

Energy Efficient Methods for Millimeter Wave Picocellular Systems

NYU Wireless

Outline

2

Millimeter Wave: Potentials and Challenges

Capacity Estimation 28 GHz Measurements in New York City

Power Consumption Issues

Subband Scheduling

Conclusions and Future Work

NYU Wireless

mmW: The New Frontier for Cellular

3

Potential 1000x increase over current cellular: Massive increase in bandwidth Near term opportunities in LMDS and E-Bands

Up to 200x total over long-time

Spatial degrees of freedom from large antenna arrays

From Khan, Pi “Millimeter Wave Mobile Broadband: Unleashing 3-300 GHz spectrum,” 2011

NYU Wireless

Key Challenges: Range

4

Friis’ Law:

Free-space path loss ∝ Increase in 20 dB moving from 3 to 30 GHz

Shadowing: Significant transmission losses possible: Mortar, brick, concrete > 150 dB Human body: Up to 35 dB

NLOS propagation relies on reflections

NYU Wireless

Challenges: Power Consumption

5

High bandwidths

Large numbers of antennas

ADC bottleneck Digital processing of all antennas not possible

Low PA efficiency in CMOS (often < 10%)

NYU Wireless

Outline

6

Millimeter Wave: Potentials and Challenges

Capacity Estimation 28 GHz Measurements in New York City

Power Consumption Issues

Subband Scheduling

Conclusions and Future Work

NYU Wireless

NYC 28 GHz Measurements

7

Focus on urban canyon environment Likely initial use case Mostly NLOS “Worst-case” setting

Measurements mimic microcell type deployment: Rooftops 2-5 stories to street-level

Distances up to 200m

All images here from Rappaport’s measurements:

Azar et al, “28 GHz Propagation Measurements for OutdoorCellular Communications Using Steerable BeamAntennas in New York City,” ICC 2013

NYU Wireless

Path Loss Comparison

8

Measured NLOS path loss in NYC > 40 dB over free-space > 40 dB worse than 3GPP

urban micro model for fc=2.5 GHz

> 20 dB over prev. studies

But, will still see large capacity gain possible

NYU Wireless

Simulation Parameters

9

Parameter Value Remarks

BS layout Hex, 3 cells per site, ISD = 200m

Similar to 3GPP Urban Micro (UMi) model (36.814)

UE layout Uniform, 10 UEs / cell

Bandwidth 1 GHz

Duplex TDD To support beamforming

Carrier 28 GHz

Noise figure 7 dB (UE), 5 dB (BS)

TX power 20 dBm (UE), 30 dBm (BS) Supportable with 8% PA efficiency

Scheduling Proportional fair, full buffer traffic

Static simulation corresponds to equal bandwidth

Antenna 8x8 2D uniform array at UE and BS)

Long-term beamforming. Single stream, no SDMA

NYU Wireless

SNR Distribution

10

SNR distribution similar to current macrocellulardeployment

But, depends on: Power Beamforming

NYU Wireless

Comparison to Current LTE

11

Initial results show significant gain over LTE Further gains with spatial mux, subband scheduling and wider bandwidths

System antenna

Duplex BW

fc (GHz)

Cell throughput (Mbps/cell)

Cell edge rate(Mbps/user, 5%)

DL UL DL UL

mmW(64x64)

1 GHz TDD

28 780 850 8.22 11.3

Current LTE(2x2 DL,2x4 UL)

20+20MHz FDD

2.5 53.8 47.2 1.80 1.94

~ 15x gain ~ 5x gain

Parameters from previous slide with 50-50 UL/DL split & 20% overhead

LTE capacity estimates from 36.814

NYU Wireless

Outline

12

Millimeter Wave: Potentials and Challenges

Capacity Estimation 28 GHz Measurements in New York City

Power Consumption Issues

Subband Scheduling

Conclusions and Future Work

NYU Wireless

RF Beamforming

13

Low power consumption Single mixer and ADC / DAC per digital stream RF phase shifting may lack accuracy

From Khan, Pi “Millimeter Wave Mobile Broadband: Unleashing 3-300 GHz spectrum,” 2011

NYU Wireless

BB Analog Beamforming

14

Intermediate power consumption One mixer per antenna and stream One DAC / ADC + BB amp per stream Lower mixer linearity requirement

NYU Wireless

Component Power Consumption

15

Component Power (mW)

RF BF Analog BF Remarks

PA * N N Typ efficiency = 8%

LNA 20 N N

RF shifter 23 KN 0

Mixer 19 K N

LO buffer 5 K 2N-1

Filter 14 K N

Phase rotator 1.4 0 KN

BB amp 5 K K

ADC 255 K K 6 bit, 2 Gsps

K=# streams, N=#antennas

NYU Wireless

Outline

16

Millimeter Wave: Potentials and Challenges

Capacity Estimation 28 GHz Measurements in New York City

Power Consumption Issues

Subband Scheduling

Conclusions and Future Work

NYU Wireless

Subband Scheduling

17

Reduce UE power consumption A/D power scales linearly with

bandwidth

Reduced peak rate to individual UE

But, no loss in total capacity in DL

Improved capacity in UL

Enables smaller MAC transport blocks.

NYU Wireless

Beamforming Optimization

18

Each UE needs to only support one digital stream

But, BS ideally uses different beams to each UE

What is possible with limited number of digital streams?

NYU Wireless

Multiple Access & Other Benefits

19

Power saving also possible via TDMA and DRX

Very inefficient in power-limited regime 10x decrease in UL

Reduced MAC Transport block Ex: 125 us TTI x 1 GHz x 2

bps/Hz = 250,000 DoF

NYU Wireless

Beamforming Optimization

20

Parameters = # antennas, # streams at BS unitary beamforming matrix ∗ = long-term SNR of UE

Utility optimization:

max

Non-convex, but can perform local optimization easily Weighted power algorithm.

NYU Wireless

Optimization Results

21

4 streams is adequate with 10 UEs per cell

Uplink Rate CDF Downlink Rate CDF

NYU Wireless

Outline

22

Millimeter Wave: Potentials and Challenges

Capacity Estimation 28 GHz Measurements in New York City

Power Consumption Issues

Subband Scheduling

Conclusions and Future Work

NYU Wireless

Rethinking LTE for mmW

23

5th Generation cellular

Many innovative technologies

Directional relayingMesh networks

Carrier aggregation for macro-diversity

NYU Wireless

Summary

24

Significant potential for capacity increase in mmW 1GHz TDD mmW offers 15x over 20+20 MHz LTE FDD But, throughput gains are not uniform

Systems appears power-limited: Heavy dependence on dense cells & beamforming Strong difference to current cellular systems Traditional methods for increasing capacity may be limited

Capacity tied closely with front-end capabilities Power consumption issues Number of digital streams, beamforming, …

NYU Wireless

References

25

Khan, Pi, “Millimeter-wave Mobile Broadband (MMB): Unleashing 3-300GHz Spectrum,” Feb 2011, http://www.ieee-wcnc.org/2011/tut/t1.pdf

Pietraski, Britz, Roy, Pragada, Charlton, “Millimeter wave and terahertz communications: Feasibility and challenges,” ZTE Communications, vol. 10, no. 4, pp. 3–12, Dec. 2012.

Akdeniz, Liu, Rangan, Erkip, “Millimeter Wave Picocellular System Evaluation for Urban Deployments”, Apr 2013, http://arxiv.org/abs/1304.3963

Azar et al, “28 GHz propagation measurements for outdoor cellular communications using steerable beam antennas in New York City,” to appear ICC 2013

H. Zhao et al “28 GHz millimeter wave cellular communication measurements for reflection and penetration loss in and around buildings in New York City,” ICC 2013

Samimi,et al “28 GHz angle of arrival and angle of departure analysis for outdoor cellular communications using steerable beam antennas in New York City,” VTC 2013.